Check the output
Sound goes to whatever output your system picked last, which is often not the device in front of you. Pick your speakers or headphones explicitly and set the volume near half.
→ Output selected on purpose, volume around 50%.
A free online speaker test for your speakers and headphones. Pick your output, choose the quick check or the full set, and press play. The quick check walks through music, each speaker, bass and treble and gives you a verdict; the full set adds single frequencies, a 20 Hz to 20 kHz sweep, phase, headphone balance, surround direction and loudness steps, with live meters proving the browser is playing. Everything runs in your browser, nothing is uploaded, and the microphone is never touched.
The browser hides device names until microphone permission is granted, and this site never asks for it. Names may stay generic, switching still works.
This slider only scales the test signal. Use your device volume keys for overall loudness, and start low before bass tones.
Play the same test signal through the left speaker, the right speaker, or both. If one side stays silent, the test tells you where the chain is broken.
Keyboard: Space plays Both, arrow keys play one side, Esc stops everything.
Alternating bursts switch between channels with an on-screen cue, so you can point at the side you hear. The pan sweep glides a tone from far left to far right.
The same 440 Hz tone in both channels, with one side inverted. Only the phase changes, so judge the centre image, not the loudness.
Single tones, a continuous generator and two sweeps. Start quiet: low tones carry a lot of energy.
Drag to any frequency: the tone follows while you slide, so you can hunt a rattle, a room resonance or the exact note. Square and sawtooth expose buzz sooner than a sine.
A 30 second logarithmic sweep. Note where the tone disappears or gets harsh. Not a hearing test and not a diagnosis. It only shows the top of what the whole chain still reproduces: file, codec, amplifier, driver, and your ears. Most adults lose the top of the range with age.
Put the headphones on, then play each side alone and both together. Both cups should sound equally loud with the same tone.
Also check the fit: a poor seal loses bass first. If one cup is quiet, swap left and right on the cable to see if the fault follows the cable or stays with the cup.
Drag until the tone sits in the middle of your head, or exactly between the two speakers. Positive means the right side is louder. Within 1 dB is inaudible; more than that is a balance setting, a placement problem or a failing driver.
The left ear gets 200 Hz, the right ear 200 Hz plus the beat you pick. The pulse you hear is the difference between the two tones, and it exists only in your head, so speakers cannot produce it.
Short bursts placed in eight directions around your head. Close your eyes and point at each sound.
Tap a direction to hear it.
System reports: − channels
A browser tab outputs stereo. This test checks spatial rendering and headphone imaging, not physical 5.1 or 7.1 routing. The channel count below the pad is what your system reports.
A 1 kHz reference tone in four calibrated steps. Each step should be clearly louder than the previous one, and both channels should match.
Steps are relative to full level: -20 dB, -12 dB, -6 dB and 0 dB. If two steps sound identical, or left and right differ, something in the chain is compressing or tilting the signal. This checks the system chain at fixed relative levels; the slider at the top only scales the test signal.
Play each item and mark what you heard. The verdict appears only when all five are marked Heard it; leaving one blank or marking Nothing gives no verdict, and neither blocks anything. Replay any item, or switch to All tests whenever you like.
Confirmed 0 / 5
All five items came through on both sides. If you still hear buzzing or distortion at high volume, go into All tests and check phase and the sweep.
The music is generated in your browser, not streamed: no file is downloaded and nothing is uploaded.
Press any play button and the channel meters appear here
If the meters move but you hear nothing, the browser is playing correctly and the problem sits in your output device, system volume or hardware.
SpeakerTester is a single page with a one-click quick check, six detailed tests and zero permissions. It exists because speaker problems show up at the worst moment, and the check itself should take a minute, not an afternoon.
Every test is built on the Web Audio API, the audio engine inside every modern browser. Tones are computed at play time, noise is filtered in memory, and the channel meters read the exact signal the page sends to your output device. Nothing is streamed, nothing is uploaded, and the microphone is never touched.
It is a listening test, not a laboratory: loudness here is relative, the sweep is not a measured frequency-response curve, and the top of the range depends on your hearing as much as your speakers. Saying the limits plainly is what makes the results worth trusting.
| Web Audio API | The W3C standard every generator, panner and analyser on this page is built on. |
|---|---|
| ITU-R BS.775-2 | The reference for stereo loudspeaker placement and listening angles. |
| IEC 60268-5 | The standard loudspeaker measurement series, useful context for what a listening test is not. |
A speaker test answers three questions: does sound come out, do both channels work, and how much of the frequency range survives the trip. The whole check takes about a minute.
Sound goes to whatever output your system picked last, which is often not the device in front of you. Pick your speakers or headphones explicitly and set the volume near half.
→ Output selected on purpose, volume around 50%.
Step 02 offers two ways in. The quick check plays music, then each speaker, then bass and treble, and asks what you heard. All tests hands you the six detailed tests and you drive each one yourself.
→ Mode selected, first sound coming from the right device.
Mark each item you heard in the quick check: five marks give a verdict. If something fails, switch to All tests and run the bass descent from 150 Hz, the full 20 Hz to 20 kHz sweep and the phase check.
→ Verdict read, bass limit found, no rattles on the sweep.
Half of all failed tests are just sound going somewhere else. These are the exact places to look on each platform.
| Platform | Where to look | What trips people up |
|---|---|---|
| Windows 11 / 10 | Settings → System → Sound, or right-click the taskbar speaker icon | The browser can be muted per-app in the Volume mixer while everything else plays fine. |
| macOS | System Settings → Sound → Output | The Output balance slider drifts off-center easily and quietly kills one side. Center it. |
| iPhone / iPad | Control Center → tap the AirPlay icon on the audio card | Flip the Ring/Silent switch if a tone refuses to play. The side buttons set the media volume. |
| Android | Settings → Sound and vibration → Media volume | The media slider also lives in the pull-down quick settings. Some phones keep routing to a paired-but-off Bluetooth device. |
| Any laptop | The output list itself | The single most common cause: the laptop still sends sound to an HDMI monitor or a sleeping Bluetooth headset. Pick "Speakers" explicitly. |
On laptops and phones, remember the physics: built-in speakers are small, so missing deep bass is expected behavior, not a fault you can fix in settings.
Every test on this page is the same five-stage chain, built directly on the Web Audio API. No plugins, no downloads, and no audio files fetched from anywhere.
A sine oscillator or a pink-noise buffer, computed live by your browser. The signal is mathematically exact, which is why it makes a fair test.
The volume slider scales the signal before it leaves the page. It is a relative control, not a calibrated level in dB.
A stereo panner routes the signal left, right or both. A spatial panner places bursts around your head for the surround test.
The meters tap the exact signal on its way out, which is why moving bars prove the browser side of the chain works.
Your audio hardware receives the signal and does the rest. Everything after this point belongs to your device.
Loudness depends on your volume setting and hardware. The tool compares channels against each other, it does not measure dB SPL.
The tone is a pure sine. Any coloration, distortion or rattle you hear comes from your chain, never from the test signal.
Meters prove the browser sent audio. They cannot prove a speaker moved air. That half of the test belongs to your ears.
The quick check answers whether sound comes out at all. The six detailed tests answer why it might not. Read together they tell you whether the fault is the browser, the connection, the speakers, or simply physics.
Both channels answer equally, the sweep runs clean and the bass reaches typical depth for the device. If it still sounds wrong, the remaining suspects are placement and the room.
Meters moving without sound, or one channel silent, points at the chain: output device, system volume, cable or jack. The table below tells you which link to pull first.
Laptop speakers roll off near 100 Hz and adult hearing tops out between 15 and 17 kHz. Some results are physics, not faults. Check the table before you shop for repairs.
| Test | What you observe | What it means | Next step |
|---|---|---|---|
| Sound test | Left and Right answer at equal loudness | The stereo path is healthy | Continue with the sweep |
| Sound test | Meters move, but silence | The browser is playing; the output side is not | Raise system volume, check device and cable |
| L/R alternating | One side never sounds | Broken channel or off-center balance | Center the balance, reseat the plug, swap cables |
| Pan sweep | A hole or dip in the middle | Out-of-phase wiring or speakers too far apart | Run the polarity check below |
| Bass descent | Fades near 100 Hz on a laptop or phone | Normal roll-off for small drivers | Use headphones or a subwoofer for deep bass |
| Bass descent | Full-size speakers give up in the hundreds | Bass driver or port problem | Compare against headphones, inspect the driver |
| Sweep | A rattle or buzz at specific notes | Loose part, tired surround or a room resonance | Note the frequency, inspect the driver and fittings |
| Sweep | Top vanishes above 15 kHz | Normal adult hearing limit | No action needed |
| Volume steps | Two adjacent steps sound identical | Compression or limiting somewhere in the chain | Disable EQ and effects, test wired instead of Bluetooth |
| Surround | Directions blur on speakers | Expected: the test renders spatial audio in stereo | Use headphones for the imaging test |
Twelve terms cover everything this page can measure and everything it talks about.
The unit of frequency: cycles per second. Human hearing spans roughly 20 Hz to 20 kHz. One kHz is a thousand cycles.
The range of frequencies a device reproduces, and how evenly. A listening sweep can reveal holes, but a measured curve needs a microphone.
Stereo carries two independent signals, left and right. Each channel can fail on its own, which is half of all speaker faults.
Where a sound appears between the speakers. A correctly imaged pair places voices exactly in the middle, between the two boxes.
Noise with equal energy per octave. It sounds balanced and is the standard signal for matching speaker pairs.
A mathematically pure single frequency. Whatever it sounds like is your equipment, never the source.
The point where small speakers lose low-frequency output. Laptops typically roll off near 100 Hz, phones even earlier.
Reversed wiring flips one speaker from push to pull, cancelling bass in the middle. The polarity check on this page catches it in a minute.
Head-related transfer functions: how your head and ears shape sound arriving from each direction. The surround test uses them to place sounds around you.
The logarithmic unit of level. The volume test uses relative steps of 6 and 12 dB, which most people hear as clearly louder.
What happens when a driver or an amplifier runs out of range: the smooth wave flattens and the sound turns harsh.
The loudness ratio between left and right. An off-center balance slider quietly kills one side and is the first thing to check.
A test signal is a known quantity. Because you know exactly what should come out, anything you hear, or do not hear, belongs to your speakers. Music cannot give you that certainty, and YouTube test videos add their own problems.
| Signal | What it is | Best for |
|---|---|---|
| Sine tone | A mathematically pure single frequency, generated live in your browser | Isolating one frequency, channel tests, the bass descent |
| Pink noise | Equal energy per octave, so it sounds balanced from top to bottom | Matching a speaker pair, hearing the whole range at once |
| Frequency sweep | A 30 second glide from 20 Hz to 20 kHz | Finding drop-outs, rattles and harsh spots in one pass |
| Your favorite song | Mastered, compressed and mixed with artistic intent | A final reality check once the technical tests pass |
| YouTube test videos | Compressed audio behind a player with ads, latency and no instant stop | Convenient in a pinch, but never as precise as a generated tone |
A track you know by heart is the fastest way to answer the question the tones cannot: does this system sound right? Use it after the technical tests, not instead of them. Good test tracks share a few traits:
Stereo carries two independent channels from the recording to your ears, and every link in that chain can fail on one side only. Three checks cover the whole path.
Playing each side alone confirms the full path: source, cable, amplifier and driver. If both sides work alone but a song feels lopsided, the recording itself is the reason.
A continuous tone gliding from far left to far right should move smoothly, with no dip in the middle. A hole in the center usually means the speakers are out of phase or set too far apart.
Reseat the plug fully, try another port and cable, and center the system balance before blaming the speaker. A mono cable in a stereo jack shorts exactly one side.
If left answers from the right and right from the left, the chain crossed itself somewhere. All three fixes take under a minute:
With an amplifier or receiver, physically swap the left and right cables at the amp end. Ten seconds and zero settings.
Players like VLC have a stereo flip, and Equalizer APO on Windows can swap the mapping system-wide. Useful when the wiring is sealed, say a soundbar or built-ins.
If the pair is identical and symmetrical, swapping their positions achieves the same result and leaves the wiring untouched.
The frequency test answers one question per tone: can this system reproduce it? Knowing the bands turns "it sounds off" into a specific, fixable observation.
| Band | Range | What lives there | Note for testing |
|---|---|---|---|
| Sub-bass | 20 to 60 Hz | Movie rumble, the deepest kick drums | Only subwoofers reach here with useful level |
| Bass | 60 to 250 Hz | Bass guitar, the body of the kick | The bass descent starts at 150 Hz for this reason |
| Low mid | 250 to 500 Hz | Warmth of vocals and guitars | Boxy, muddy sound usually means this range is boosted |
| Mid | 500 Hz to 2 kHz | Vocal intelligibility, snare body | The most important range for speech |
| Upper mid | 2 to 4 kHz | Vocal presence, attack of instruments | Where the human ear is most sensitive |
| Presence | 4 to 6 kHz | Clarity, consonants, string detail | Harsh when boosted, dull when cut |
| Brilliance | 6 to 20 kHz | Cymbals, air, shimmer | Roll-off with age is completely normal |
A driver must move air to make bass. Laptop speakers typically lose output below about 100 Hz and cannot reproduce 40 Hz at all. Missing bass on a laptop is physics, not a fault.
Young ears reach about 20 kHz. Most adults top out between 15 and 17 kHz, and it drops with age. Losing the top of the sweep is expected, not a problem.
This is a listening test, not a calibrated frequency-response chart. It reveals drop-outs, rattles and hard limits. Trust a measured curve from a microphone for anything scientific.
Surround sound means different things in a browser and in a living room, and the difference matters for what you can actually test.
A browser tab plays stereo: two channels, left and right. Physical 5.1 and 7.1 routing, meaning front center, surround back and the LFE channel, is decided by your operating system and driver before any website gets involved. No web page can certify that routing.
What a browser can do is spatial rendering. On headphones, the surround test above places short bursts in eight directions around your head using head-related transfer functions. That is a genuine imaging test for headphones, and a pleasant demo on speakers, but it is not channel certification.
Work down this table in order. The first two rows separate a browser problem from a hardware one, which is the split the meters in the test give you for free.
| Problem | What you notice | Likely cause | Fix |
|---|---|---|---|
| No sound at all | Meters do not move | Browser tab muted or wrong output device | Right-click the tab icon in Chrome to unmute; check the system output list |
| No sound, meters move | Signal confirmed playing | Output device off, system volume down, broken connection | Raise system volume, check power and cables, try another output |
| One channel silent | Left or right alone works | Balance slider, half-seated plug, mono cable | Center the balance, reseat the plug fully, test with another cable |
| No bass | Descent test dies early | Natural roll-off, subwoofer off, crossover set wrong | Expect laptop roll-off near 100 Hz; power the sub, crossover 80 to 120 Hz |
| Distortion at volume | Harshness on the sweep | Small drivers overdriven, EQ boosts stacking up | Turn it down and switch off bass boost; if it persists quietly, suspect the driver |
| Bluetooth sounds thin | Weak, noisy audio | Headset mode (mic plus speakers) forces a low-quality codec | Disable call audio for that device or test with the microphone idle |
| Crackle and drop-outs | Sound cuts mid-tone | Wireless interference, low battery, damaged cable | Recharge, shorten the distance, or test with a wire |
A wiring polarity error reverses one speaker, and the two sides then cancel each other in the bass. The check is quick:
Speakers fail from mistreatment far more often than from age. A few habits prevent almost all of it.
After long idle, bring speakers up from low volume. Amplifier thump and full-level test tones are how surrounds crack and voice coils burn.
Powered speakers need airflow around their vents. Dust the grilles gently; blocked vents cook amplifiers slowly.
Pull plugs by the connector, coil cables loosely and keep jacks clean. Crackling that follows the cable is the cable.
Keep speakers off sun-baked windowsills and away from amplifier heat. Stability matters too: a wobbling stand buzzes at exactly one bass note.
Half of every video call runs in the other direction. If your speakers passed today and people still cannot hear you, the microphone is the next link in the chain.
MicTester is our sister site with the same privacy promise: the test runs entirely in your browser and any recording stays on your device.
Twelve questions that cover almost everything a speaker test can and cannot do.
Start speaker test ↗Open the speaker test at the top of this page, set the volume near half and run the quick check: music, left, right, bass and treble. Mark each item you heard; five marks give you a verdict. Then switch to All tests for the bass descent and the sweep to see how much of the range your speakers actually reproduce.
No, if you start at moderate volume. The tones are ordinary audio signals and every continuous tone stops automatically after 30 seconds. Deep bass carries more energy, so lower the volume before the low frequencies and keep the sweeps short.
Check three things in order: the system balance slider, which drifts off-center easily; the seating of every plug in the chain; and the cable itself. If one speaker stays silent on every source, the fault is inside the speaker or its amplifier.
Small drivers cannot move enough air to reproduce low frequencies. Laptop speakers typically roll off around 100 Hz and phones even earlier. That is physics, not a fault. Full-size headphones or a subwoofer are the only real fixes.
A browser tab outputs stereo, so no website can verify physical surround routing. Use the surround test here for spatial imaging on headphones, and your operating system or receiver test tones for channel-by-channel 5.1 and 7.1 verification.
Yes. Young ears reach about 20 kHz, but the top drops with age and most adults hear 15 to 17 kHz at best. Losing the top of the sweep is expected. Only a professional audiometer can actually measure hearing.
Channel tests usually use 440 Hz, the concert pitch A, because every speaker reproduces it comfortably. Volume calibration uses 1 kHz, where the ear is most sensitive. The bass descent steps from 150 Hz down to 20 Hz.
Yes, in any modern mobile browser: Safari on iOS, Chrome on Android. On iPhone, flip the Ring/Silent switch if a tone refuses to play, and remember the side buttons control the media volume.
Center the balance slider first. Then swap the channels: a difference that follows the cable is a connection fault, while one that stays with the same speaker or cup is hardware. On headphones, a poor seal on one side quietly kills bass first.
The speaker test is free and needs no account. All audio is generated locally in your browser, nothing is recorded, and the microphone permission is never requested or used.
Tones first, music second. A tone is a known quantity, so anything you hear is your speaker and not the mix. Music is mastered and compressed, which hides problems. But a familiar track remains the best final check of whether the system sounds right overall.
Unmute the browser tab, which Chrome marks with a speaker icon, confirm the system output points at the right device, and raise the media volume. Those three steps fix most failed tests before any hardware is touched.
One quick check or six detailed tests, one page, one minute. Check your speakers and headphones now, then send the microphone to our sister site.